Harness the power of AI with DeepSeek V4-Pro in both chat and reasoning modes by creating requests to Together.ai's API with just a few lines of code!

DeepSeek in Python

Your first DeepSeek request!

Making requests to DeepSeek models

Building a DeepSeek request

Specifying a DeepSeek model

Digging into the response

Enabling reasoning with DeepSeek

Getting the model to think

Tuning reasoning effort

Introduction to DeepSeek

DeepSeek's models have created headlines the world over. In this chapter, you'll generate and refine text, classify content with AI, analyze sentiment, and explore shot prompting to optimize response outputs. Finally, you'll enable DeepSeek V4-Pro's reasoning mode for code debugging and analyzing legal documents.

Summarizing and editing text

Find and replace

Text summarization

Calculating the cost

Text generation

Content generation

Generating a product description

Shot prompting

Zero-shot prompting with reviews

One-shot prompting: will it be enough?

Few-shot prompting: all the examples!

Deep analysis and code generation

Coding with reasoning mode

Analyzing legal contracts with reasoning mode

Prompting DeepSeek Models

Time to create your very own AI-powered chatbot! You'll learn how to write effective system messages, provide examples in the form of user and assistant messages, and integrate a message history to retain context between prompts. You'll finish by building a multi-turn coding chatbot powered by DeepSeek V4-Pro.

Chat roles and system messages

Utilizing systems messages

Adding guardrails

Utilizing the assistant role

Chat messages

Structuring chat messages

Adding assistant messages

More assistant messages!

Multi-turn conversations with DeepSeek

Creating a conversation history

Creating an AI chatbot

Creating a reasoning chatbot for coding

Congratulations!

Building Conversations with DeepSeek

Get started using DeepSeek V4-Pro in both chat and reasoning modes to build inspiring customer experiences and automate business processes. In this course, you'll use DeepSeek models, via Together.ai's API, to solve a wide range of real-world tasks, including generating content for marketers and e-commerce, summarizing customer conversations for support teams, and building a chatbot to help people code in Python. Learn what all the DeepSeek hype was about today!

<h2>Discover How to Access DeepSeek AI Models</h2>Since the release of its V3 and R1 models, DeepSeek has emerged as a market leader in developing high-performance, open-source large language models (LLMs) for the AI community. These models were developed with cutting-edge research that reportedly slashed training costs by up to a factor of 30 compared to similar market-leading models at the time.<br><br> To begin extracting business value from DeepSeek's models, you'll need to learn to work with their Application Programming Interface, or API. Throughout the course, you'll learn to use Python code to send requests to use DeepSeek's models. Discover what the DeepSeek hype was all about!<br><br> <h2>Solve Real-World Problems Using DeepSeek Models</h2> DeepSeek models have a wide range of potential applications. In this course, you'll use DeepSeek to generate and transform text content, perform sentiment analysis, and categorization.<br><br> <h2>Begin Developing AI Applications</h2> As AI becomes more accessible to individuals and organizations, the bar for developing AI-driven product features and automations continues to fall. Right now, every organization has the capability to begin using AI to streamline business processes and develop more personalized customer experiences. DeepSeek models are accessible via APIs so software engineers and developers can start integrating AI into applications and systems.

Intermediate Python

Discover what all of the DeepSeek hype was really about! Build applications using DeepSeek's R1 and V3 models.

Working with DeepSeek in Python

{"role": "system",
 "content": "You are a patient and informative teacher."}

{"role": "user",
 "content": "I'm really struggling with a coding problem."}

{"role": "assistant",
 "content": "Don't worry, we'll get you sorted! How can I help?"}

{"role": "user",
 "content": "How do I access dictionary values using their keys?"}

Structuring chat messages

Exercice interactif pratique